RE: ALT text survey

<<LRK:: Thus people still run into this problem if their screen reader
doesn't turn on the flag.  Do all screen readers in use today turn on
this flag?  I think not, and as in interim practical measure it's
desirable to accommodate folks using those readers.
plus
sighted people who have images off just for speed would benefit from
seeing the alt text
plus
not everyone has IE.>>

Nonetheless, our guidelines should not work around "bugs" (AT&T's term) in
browsers.  If we do, we get into this argument of "what is the worst popular
browser" and go from there.

<<Ah, I see the misunderstanding.  The our guidelines do not say to omit
height and width.  In fact, they absolutely require height and width.
What they suggest is to try to make the height and width large enough so
that the ALT text shows.>>

I'm not saying that it should be eliminated (it won't be, it's too useful),
but it's silly to think we can tell people how big their images should be.

Looking at http://www.att.com, HEIGHT is used with less than 24 points 7
times out of 15.  Also, the numbers are arbitrary - what if the font size is
larger?  Then 24 points isn't enough.  That only means that the text isn't
clipped vertically, what if the ALT text is greater than a few words, it
might be clipped horizontally as well.  Given this logic, we should have
minimum WIDTH as well, based on a metric of the size of the text, etc.

People create small images, they will wonder why the W3C says not to.  We
will say "because old graphical browsers can't display the ALT text."  They
will say "Upgrade to something that works!"

Guidelines should not work around browser implementation.  I will say that
over and over and over again until it sinks in.

In order for the guidelines to gain acceptance, there has to be a clear
understanding of the problem and that has to be grasped by the HTML author
(who will certainly be using the latest tools him/herself).

If we load up the guidelines with exceptions and such, the author will not
be as understanding to the problem and less likely to make a change.  Lets
make this document incredibly lightweight and useful rather than a series of
workarounds.


-----Original Message-----
From: Kasday, Leonard R (Len), ALTEC [mailto:kasday@att.com]
Sent: Wednesday, February 11, 1998 1:52 PM
To: 'w3c-wai-gl@w3.org'
Subject: RE: ALT text survey



chuck...
thanks for the better explanation of alt text, but I think the original
>point still holds.  You wrote:
>
>> With IE3 and IE4, the bounding box of a image is expanded to include all
>>the
>> ALT text when "Show Images" is turned off and when the
ScreenReaderPresent
>> flag is turned on by a screen reader.

LRK:: Thus people still run into this problem if their screen reader
doesn't turn on the flag.  Do all screen readers in use today turn on
this flag?  I think not, and as in interim practical measure it's
desirable to accommodate folks using those readers.

plus

sighted people who have images off just for speed would benefit from
seeing the alt text

plus

not everyone has IE.
>
>> With IE4.01, in addition to the conditions above, the feature can be
>> manually turned on from the "Always expand ALT text for images" option in
>> the Advanced tab of the Internet options dialog.

LRK:: that is a very good feature.

> 
>> <<It's not true that this must radically affect page layout.>>
>
>> It does affect page layout, regardless of borders, etc.  HEIGHT and WIDTH
>> determine how much space is allocated for the image before it's
downloaded.
>> If it's not specified, then the text that flows around it will be
>> reformatted after the image is displayed.  This can cause unsightly
>>flashing
>> of the screen when there are a lot of images being downloaded and
>>displayed.

LRK::
Ah, I see the misunderstanding.  The our guidelines do not say to omit
height and width.  In fact, they absolutely require height and width.
What they suggest is to try to make the height and width large enough so
that the ALT text shows.   


Len

>
>All opinions expressed here are my own, not necessarily those of my
employer.
>=============================================================
>kasday@att.com         phone 732 949 2693
>
>Leonard R. Kasday
>Room 1J-316A
>AT&T Laboratories
>101 Crawfords Corner Rd.
>Holmdel NJ 07733
>

Received on Wednesday, 11 February 1998 19:16:20 UTC